Combining EEG signals from the 2 members of a team to improve event identification\u2606

Metaverse-based educational environments exhibit highly dynamic and interactive learning processes in which learners' ethical awareness evolves through latent developmental stages and abrupt behavioural shifts. Traditional static assessment approaches fail to capture such temporal dynamics. To address this limitation, this study proposes a hybrid Bayesian sequential modelling framework for dynamic evaluation of ethical awareness, using fully simulated longitudinal behavioural sequences designed to mimic realistic learning patterns in metaverse-based educational environments. The framework integrates Kalman filtering, Bayesian Online Change-Point Detection (BOCPD), and a duration-aware Hidden Semi-Markov Model (HSMM) to construct a reproducible time-series modelling pipeline. Kalman filtering is employed to estimate smoothed latent ethical trajectories, BOCPD identifies structural transitions triggered by behavioural fluctuations, and HSMM models developmental stages by incorporating explicit state-duration distributions, thereby reducing spurious short-term transitions. In addition, an interrupted time-series model with propensity score matching (ITS + PSM) is used to estimate the causal effects of instructional interventions.Experimental results demonstrate that the proposed hybrid Bayesian model sensitively detects key change points within metaverse learning contexts, yields stable stage-inference results, demonstrates the methodological capability of the proposed framework to support future intervention analysis. This work provides an interpretable, traceable, and reproducible methodological approach for evaluating ethical awareness in dynamic digital learning environments and offers valuable implications for ethical governance in metaverse-based education.

ABSTRACT Human-robot collaboration (HRC) technology can facilitate high-variety, low-volume (HVLV) assembly systems. However, simultaneously empowering both the operator and the cobot to adopt changes remains challenging. This study proposes a bidirectional HRC empowerment model to achieve high team fluency in the HVLV assembly scenario. Deep learning-based augmented reality (AR) assembly assistance is offered to empower operators for a wide range of assembly tasks, and a hidden semi-Markov model (HSMM)-based prediction is proposed to enhance the robot’s cognitive capability in relation to the operator. An AR and HSMM empowered HRC assembly station, featuring two agents (the operator and the cobot), is implemented using the current model. The team fluency metrics and the NASA-TLX mental workload instrument are measured for four typical assembly scenarios: manual assembly, basic HRC, HSMM-assisted HRC, and HSMM and AR-assisted HRC assembly. The experiment results confirm that the proposed model increases assembly efficiency by 13% and decreases the overall workload by 9.6%. The proposed bidirectional HRC empowerment model, which integrates AR technology and the HSMM algorithm, contributes to both fluency and mental workload in diverse HRC assembly scenarios within the HVLV systems.

HighlightsWhat are the main findings? Participants that completed a guided mindfulness session following an acute stressor spent more time in a brain state in which the salience network was more active.Following an acute stressor, participants in the control group spent more time in brain states in which the default mode network was more active.What are the implications of the main findings?Mindfulness may work to shift the brain out of states responsible for rumination and into a state that better supports emotional regulation and recovery following stress.This work offers a novel approach to testing and optimizing mindfulness-based therapies.Background: Previous research has found that mindfulness-based techniques are beneficial for reducing stress in heavy-drinking individuals. However, the underlying neurobiology of these stress-reducing effects are unclear. Moreover, much of the research examining neurobiological correlates of mindfulness has used static functional connectivity, suggesting that brain activity goes unchanged for the entire length of an MRI scan. Methods: In the current study, we used a state-based dynamic functional connectivity model to examine brain states during either a 10 min mindfulness session or resting control that followed an individually tailored stress imagery task. Using a hidden semi-Markov model (HSMM), six brain states and the associated dynamics of state traversal were estimated for a population of moderate-to-heavy drinkers (N = 32). We modeled the 36 Schaefer atlas regions spanning the salience and default mode networks, and the HSMM characterized each state by its distinct multivariate pattern of activity and covariance structure. Group differences in dwell times, transition behavior, and overall state dynamics were evaluated using permutation tests and mixed-effects models. Results: Participants that experienced the mindfulness session had more transitions and longer time spent in states in which the salience network was more active. Participants assigned to the control group had more transitions and increased time spent in states in which nodes of the default mode network were more active. Moreover, for control participants, increased occupancy time to SN-dominant states was associated with lower perceived stress. Conclusions: Using HSMM provided a unique insight into network connectivity during mindful states; we believe it offers a novel approach to testing and optimizing mindful-based therapies.

Analyzing the spontaneous activity of the human brain using dynamic approaches can reveal functional organizations. The co-activation pattern (CAP) analysis of signals from different brain regions is used to characterize brain neural networks that may serve specialized functions. However, CAP is based on spatial information but ignores temporal reproducible transition patterns, and lacks robustness to low signal-to-noise rate (SNR) data. To address these issues, this study proposes a new CAP framework based on hidden semi-Markov model (HSMM) called HSMM-CAP analysis, which can be performed to investigate spatiotemporal CAPs (stCAPs) of the brain. HSMM-CAP uses empirical spatial distributions of stCAPs as emission models, and assumes that the state sequence of stCAPs follows a semi-Markov process. Based on the assumptions of sparsity, heterogeneity, and semi-Markov property of stCAPs, the HSMM-CAP-K-means method is constructed to infer the state sequence and transition parameters of stCAPs. In addition, HSMM-CAP provides the inverse relationship between the number of states and sparsity. Simulation studies verify the performance of HSMM-CAP at different levels of SNR. The spatiotemporal dynamics of stCAPs are also revealed by the proposed method on real-world resting-state fMRI data. Our method provides a new data-driven computational framework for revealing the brain spatiotemporal dynamics of resting-state fMRI data.

In this paper, a novel hidden semi-Markov model (HS-MM) in the continuous-time domain is proposed, and the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H</i><sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> static output feedback control problem for hidden semi-Markov jump linear systems (HS-MJLSs) is investigated. Data transmission via a wireless network medium in HS-MJLSs is vulnerable to cyber attacks. Therefore, we adopt a resilient control strategy to ensure the stability of HS-MJLSs under hybrid cyber attacks, factoring in both denial-of-service attacks and false data injection attacks. In comparison with the conventional Markov jump linear systems, the HS-MM proposed by us is more universal, which is reflected in the following two aspects: (i) The system mode is considered to be undetectable. The jumping of the controller depends on the observation mode, and is asynchronous with the jumping of the system mode. (ii) Both the mode jumps of the original system and controller are considered to be related to the sojourn-time, which is more reasonable than the memoryless jumping processes. By using the established HS-MM and a Lyapunov-based approach, some sufficient conditions are obtained to ensure the stochastic stability of the HS-MJLSs with an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">H</i><sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> performance and hybrid cyber attacks. Finally, the proposed theoretical results are demonstrated by two simulation examples.

Background: Cardiovascular diseases remain the leading cause of global mortality (17.9 million deaths annually). Early detection through cardiac auscultation offers critical intervention opportunity, yet diagnostic accuracy varies widely (sensitivity 32-44% for clinically significant valvular heart disease). AI-powered phonocardiogram analysis presents a promising approach to democratize cardiac screening. Objective: To develop and conduct preliminary validation of HeartSound AI, a comprehensive deep learning platform for automated PCG analysis designed for smartphone deployment. Methods: We designed a modular pipeline integrating: (1) adaptive preprocessing with bandpass filtering (20-800 Hz); (2) signal quality assessment; (3) heart sound segmentation using Hidden Semi-Markov Models; (4) spectro-temporal feature extraction (MFCCs, CWT, Mel-spectrograms); (5) hybrid CNN-RNN classifier with attention; and (6) calibrated probability fusion. Development followed Good Machine Learning Practices guidelines. Evaluation utilized CirCor DigiScope dataset (5,272 recordings) and pilot clinical recordings (n=15). Results: On development data, the platform achieved weighted accuracy of 74.2% (95% CI: 71.8-76.6%) with murmur-present sensitivity of 81.3%. Heart sound segmentation demonstrated S1/S2 detection rates exceeding 94%. Real-time analysis was achieved in <2 seconds on smartphone hardware. In pilot demonstrations, 2 of 12 acceptable-quality recordings showed findings correlated with clinical evaluation. Conclusions: HeartSound AI demonstrates technical feasibility for smartphone-based cardiac screening. Prospective multi-center clinical trials are required to establish diagnostic accuracy before clinical deployment.

ABSTRACTAlcohol misuse is a significant public health concern, yet little is known about the neural dynamics associated with habitual heavy drinking, particularly during abstinence. The Triple Network Model, comprising the salience network (SN), default mode network (DMN), and central executive network (CEN), provides a framework for understanding large‐scale brain network dysfunction associated with heavy alcohol use. Using resting‐state fMRI and a Hidden Semi‐Markov Model (HSMM), we examined dynamic brain state changes in moderate‐to‐heavy drinkers (n = 38) across two conditions: typical drinking and alcohol abstinence. Our findings revealed six distinct brain states, with significant differences in state occupancy, transitions, and duration between drinking conditions. Abstinence was associated with decreased time spent in a DMN‐dominant state, a lower probability of transitioning to a state with high SN activation, and more frequent but shorter durations in a state without a distinct dominant network. These results suggest alcohol abstinence alters the temporal dynamics of these brain networks, potentially disrupting attention shifting and cognitive control mechanisms that may contribute to relapse risk. Understanding these neural adaptations will provide critical insight into the neurobiology of habitual heavy drinking and inform potential targets for future interventions.

By implementing human-robot collaboration (HRC), cobots and operators can leverage their respective strengths during production to enhance efficiency and adaptability. However, coexistence with cobots can induce psychological stress in operators, potentially impairing their performance. Therefore, monitoring operators’ psychological safety and accurately gauging their stress perception levels is crucial for cobot to respond in real-time as facing operator dynamics. This study develops an operator stress perception model by detecting their facial emotions and hand movement speeds within the HRC environment. The proposed model operates within a HRC scenario, utilizing MediaPipe for the detection of operators’ hand movement speeds and Deepface for facial expression recognition. A hidden semi-Markov model is developed to predict the operator’s perceived stress, allowing the cobot to adjust its movement speed and dwell time based on current stress perceptions. Experimental results indicate that our proposed method effectively reduces operator workload and enhances productivity.

Previous research has found that mindfulness-based techniques are beneficial for reducing stress in heavy drinking individuals. However, the underlying neurobiology of these stress-reducing effects are unclear. Moreover, much of the research examining neurobiological correlates of mindfulness have used static functional connectivity, suggesting brain activity goes unchanged for the entire length of an MRI scan. In the current study, we used a state-based dynamic functional connectivity model to examine brain states during either a 10-minute mindfulness session or resting control that followed an individually tailored stress imagery task. Using a Hidden Semi-Markov Model (HSMM), six brain states and the associated dynamics of state traversal were estimated for the population. Participants that experienced the mindfulness session had more transitions and longer time spent in states in which the salience network was more active. Participants assigned to the control group had more transitions and increased time spent in states in which nodes of the default mode network were more active. Moreover, for control participants, increased occupancy time to SN-dominant states were associated with lower perceived stress. Using HSMM provided unique insight into network connectivity during mindful states; we believe it offers a novel approach to testing and optimizing the content of mindful-based therapies.

The well-established methodology for the estimation of hidden semi-Markov models (HSMMs) as hidden Markov models (HMMs) with extended state spaces is further developed. Covariate influences are incorporated across all aspects of the state process model, in particular regarding the distributions governing the state dwell time. The special case of periodically varying covariate effects on the state dwell-time distributions — and possibly the conditional transition probabilities — is examined in detail. Important properties of these models are derived, including the periodically varying unconditional state distribution as well as the overall state dwell-time distribution. Simulation studies are conducted to assess key properties of these models and provide recommendations for hyperparameter settings. A case study involving an HSMM with periodically varying dwell-time distributions is presented to analyse the movement trajectory of an Arctic muskox, demonstrating the practical relevance of the developed methodology.

This paper addresses the fault detection problem for hidden semi-Markov jump systems with partially observed modes, which is a challenging yet critical issue in systems with dynamic and uncertain mode observations. Unlike standard semi-Markov models, hidden semi-Markov models allow for more complex dynamics but introduce difficulties in analysing mode-dependent behaviours due to unobservable modes. To tackle this, a partially mode-dependent fault detection filter is developed, incorporating both mode-independent and mode-dependent parameters, thereby accommodating extreme cases where observed modes are either always lost or consistently available. Furthermore, a novel dynamic event-triggered mechanism is proposed, featuring an adjustable parameter to balance fault detection performance and system transmission load. The filter design is guided by a set of derived inequality conditions to compute the filter gains. Numerical simulations validate the proposed approach, demonstrating its effectiveness and robustness in handling diverse scenarios with varying levels of observability and transmission constraints.

The extensive deployment of wireless infrastructure provides the possibility of locating mobile users in indoor environments using received signal strength (RSS). One approach to localization in this context is the use of Wi-Fi RSS fingerprinting and this formulation has been found to work reasonably well for location recognition of mobile phone users. For such, machine learning techniques such as hidden Markov models (HMMs) and hidden semi-Markov models (HsMM) have been extensively used to study human mobility and movements which permits the inclusion of prior knowledge about the geography of the environment alongside RSS measurements in the estimation process. Conventional HMMs, with their assumption of the Markov property, whose memory length is 1, i.e., dependency only on the last state, offer a simpler and more computationally manageable framework. Movements of typical mobile users through the quantized cells of the building, however, are usually not Markovian. HMM estimates suffer from ambiguity recognition on the movement of a Markov chain between subsets of state spaces. HsMMs outperform HMMs by allowing a semi-Markov chain with a variable sojourn time for each state, however still fail to capture the longer dependency beyond just the previous state. Combinatory Categorial Grammar (CCG) was designed to deal with the long-range dependencies in computational linguistics. In this article, we investigate the feasibility of implementing CCG as an alternative to HMM to formulate the building layout to a category of semantics and construct a walking path by CCG parsing from the RSS observations. The CCG parser allows the construction of the estimated path by recursively combining different path segments, thus building up a longer dependency between locations. The authors believe this is the first application of computational linguistics in the field of localization. Field test results demonstrate the effectiveness and reliability of the grammar-based approach which can achieve a resolution of 87.5% room-level matching accuracy based on crowdsourced fingerprints under a real large-scale university public wireless sensor network. Comparison between HMM, HsMM, and the grammar approach has been made to reveal the fact that both methods show promising performance, while the grammar approach is more reliable as HMM/HsMM can occasionally fail due to ambiguity recognition while the grammar approach consistently maintains good localization accuracy.

Portable ECG and PCG wireless acquisition system and multiscale CNN feature fusion Bi-LSTM network for coronary artery disease diagnosis.

BackgroundThe human brain is a highly interconnected and dynamic system. The study of neuroimaging indicators of future teen drinking has primarily focused on the activation of individual brain regions. We applied novel methodology to identify relationships between functional brain network dynamics and future drinking outcomes in non/low drinking teens.MethodsResting‐state functional magnetic resonance imaging (fMRI) time series from 17‐year‐old non‐/low drinking participants (n = 295) of the National Consortium on Alcohol and NeuroDevelopment in Adolescence (NCANDA) study were used to fit a Hidden semi‐Markov Model (HSMM). Regions of the default mode network (DMN), salience network (SN), and central executive network (CEN), collectively known as the Triple Network, were included in modeling. The HSMM identified each participant's most likely brain state sequence through five brain states. Poisson regression models assessed relationships between occupancy time in brain states and future drinking frequency/intensity. Sex differences were assessed with permutation testing and interaction terms in regression models.ResultsNo sex differences in network dynamics were observed. However, the relationship between occupancy times and future drinking frequency differed by sex for three brain states. Occupancy time in a state characterized by high activation in the DMN and SN, but low activation in the CEN, was negatively associated with future drinking in both sexes.ConclusionsBrain network dynamics may be useful neural markers of teen drinking predisposition. Brain dynamics that make teens vulnerable or resilient to drinking may differ between sexes.

Background:Childhood disruptive behavior problems are linked to aberrant integrity within large-scale cognitive control networks. However, it is unclear if transitory or dynamic variation in the functional brain architecture is a marker of disruptive behavior problems. The current study tested whether functional connectivity across dynamic networks is distinctly associated with the transdiagnostic symptom domain of disruptive behavior problems in children.Methods:Participants were aged 9–10 years from the Adolescent Brain Cognitive Development (ABCD) Study, who completed resting-state fMRI (N=877). We employed a dynamic connectivity approach leveraging a hidden semi-Markov model (HSMM) to identify transient properties of brain networks and states. Models estimated the time spent in each state (occupancy time) and the number of consecutive timepoints in a state (dwell time) for each participant. Linear regression models were utilized to identify distinct associations between dynamic properties (occupancy and sojourn times) and severity of disruptive behavior problems, accounting for other commonly co-occurring symptoms.Results:Dynamic network markers of disruptive behavior problems included increased time in network states characterized by globally aberrant connectivity patterns in circuitry involved in cognitive control including frontoparietal and dorsal attention networks. Replication of findings was found in a held-out sample of resting-state fMRI runs in which greater severity of disruptive behavior problems was uniquely linked to greater occupancy time in similarly characterized brain states.Conclusion:Transdiagnostic, dynamic resting-state markers of disruptive behavior problems in youth may assist in the development of brain-based biomarkers for monitoring treatment outcomes, assessing circuit target engagement and informing clinical decisions.

This article focuses on the dynamic quantized control for Takagi–Sugeno fuzzy semi-Markov jump systems (T–S FSMJSs) under fading channels and deception attacks, employing an improved event-triggered mechanism (ETM) strategy. Specifically, a more generalized semi-Markov process (SMP) that is governed by a higher-level deterministic switching signal (HLDSS) is addressed. A novel dynamic ETM is skillfully developed based on the information of quantization and two internal dynamic adjusting variables to further enhance the network bandwidth utilization. The dual asynchronous phenomenon between the plant and controller (asynchronous modes and mismatched premise variables) is addressed. This implies that the designed controller is not required to share the same membership functions and modes as the plant, establishing a more rational structure. The hidden semi-Markov model (HSMM) is attained to characterize the stochastic varying channel fading amplitudes. The Lyapunov function incorporating mode and fuzzy information is constructed to establish sufficient criteria ensuring the strictly dissipative performance and mean-square exponential stability (MSES) of the resulting closed-loop systems, and a new security fuzzy dual asynchronous controller is then developed. Finally, the efficacy and applicability of the results are demonstrated through numerical and practical examples.

In human behavior recognition using machine learning, model performance degrades when the training data and operational data follow different distributions which is a phenomenon known as domain shift. This study proposes a method for domain adaptation in the hidden semi-Markov model (HSMM) by modifying only the emission probability distributions. Assuming that the state transition probabilities remain unchanged, the method updates the emission probabilities based on the posterior distribution of the target domain. This approach enables domain adaptation with minimal computational cost without requiring model retraining. The effectiveness of the proposed method was evaluated on synthetic time-series data from different domains and actual care work data, achieving recognition performance comparable to that of models retrained for each domain. These findings suggest that the proposed method applies to various time-series data analysis tasks requiring domain adaptation.

Statistical Learning of Robotic Demo Trajectories Based on Multicriteria Segmentation and Multi-Demonstration Alignment (HSMM) addresses complex tasks in human-robot interaction and intelligent manufacturing. The research goal of this study is to automatically extract generalized key segments from multiple robotic demonstration trajectories in the absence of prior annotations and establish statistical and parametric models for universal trajectory reproduction across diverse tasks and conditions. To achieve this, the research tasks include multicriteria segmentation (speed, curvature, acceleration, direction change), trajectory alignment using Hidden Semi-Markov Models (HSMM), and subsequent implementation of statistical representations (ProMP, GMM/GMR, DMP). The proposed methodology begins with the smoothing of raw data and the identification of key points via topological simplification and non-maximum suppression, then, using HSMM, it ensures consistent segmentation of multiple demonstrations into characteristic segments. The conducted experiments confirm the results of the approach, demonstrating low reconstruction error while simultaneously improving data compression and preserving key actions, indicating the high efficiency of the method. Finally, the novelty and practical significance of this study can be highlighted by the potential industrial applications (such as welding, painting, etc.), as well as the future prospective expansions of the method to more dynamic and non-stationary scenarios, requiring adaptive and statistically grounded trajectory planning.

While various models in the Markov family have been applied to manpower system analysis, these models typically overlook the distribution of durations spent in unobservable (hidden) states within the manpower system. This study introduces a hidden semi-Markov model (HSMM) framework tailored for manpower system analysis, with a focus on incorporating the random durations of stay in hidden states. By employing the expectation-maximization (EM) algorithm, key model parameters, including the probabilities of employee transitions between states, emission probabilities, and the duration distributions for each state are estimated. The proposed method is validated using academic manpower data from a Polytechnic system in Nigeria. The results demonstrate the effectiveness of the model in capturing the dynamics of manpower transitions, offering valuable insights for improving workforce planning in hierarchical manpower system.